Prostate cancer is a disease of increasing occurrence and lethality in the American population. Moreover, this disease will have an even greater impact on our public health policies because of the enormous costs that will be involved in treating an indistinguishable patient population having a high percentage of indolent cancers. Our current theories of carcinogenesis would propose that the various developmental stages of prostate cancer might be viewed as a series of genetic lesions, which cooperate to alter the physical behavior of the prostate cell so that it becomes more rapid growing, more metastatic, more resistant to therapies. In the experiments described in this proposal, we will: 1) Survey a large number of primary human prostate cancers for the occurrence of multiple genetic lesions. Molecular analytical techniques will allow us to document the incidence of mutated or deleted p53, deleted nm23, and alterations of genes on the q arm of chromosome 10 in individual cancers. These same tissues will be examined by immunohistochemical procedures to determine whether ERB-1, ERB-2 (neu), or TGF-alpha are overexpressed and whether the rb protein is deficient in Prostate cancer cells. Genetic aberrations will be correlated with the stage of tumor development in an attempt to match sets of mutations with the developmental stage of the prostate cancer. 2) In an attempt to identify the source of genetic instability in prostate cancer, we will additionally survey these same tissues for the presence of human papillomaviruses and whether prostate cancer might overexpress the protein encoded by bcl-2, a suppressor of apoptosis. 3) Create a transgenic mouse model of prostate cancer through which progression must occur with additional genetic mutations. A prostate-specific gene promotor will be utilized to drive intense expression of dominant-mutated p53, papillomavirus or bcl-2 in the prostate gland of transgenic mice. Prostate cancers that develop in these mice will be characterized for coordinate genetic lesions to document the existence of coordinate molecular pathways to prostate cancer.